What this calculator works out
This calculator applies the classic rule of thumb for body cooling: divide the drop from an assumed starting core temperature by a fixed cooling rate to get an elapsed time in hours.
It is the simplest possible model of algor mortis, and it is included because seeing the simple version is the fastest way to understand why the real methods are more complicated. These tools are teaching aids. They are not validated for casework, and no result from them should appear in a statement, report or investigation.
Enter temperatures
Before you read the result
- The assumed starting temperature of 37°C is a population average. Normal core temperature ranges roughly 36.2 to 37.5°C between individuals and across the day, and fever or hypothermia move it much further.
- The default rate of 0.83°C per hour is the metric form of the traditional 1.5°F per hour figure. It is an average, not a constant.
- The calculator requires the measured temperature to sit between ambient and the starting value. Outside that range the model has nothing to say.
- A rectal or deep core measurement is what these models assume. Surface and oral temperatures behave differently.
Body temperature does not begin falling at a steady rate immediately. For roughly the first half hour to three hours there is a plateau during which the core barely cools, because heat is still being generated and the body's insulation resists loss. A linear model ignores this entirely, which is why simple cooling estimates systematically underestimate short intervals.
How this calculator works
A single division:
Hours elapsed = (assumed starting temperature − measured core temperature) ÷ cooling rateEvery one of the model's assumptions is questionable in a real case: that cooling is linear, that it began immediately, that the starting temperature was normal, and that a single rate applies regardless of body mass, clothing, air movement or surface.
Real cooling follows a sigmoid curve — a plateau, then a steep middle section, then a long tail as the body approaches ambient. Newton's Law of Cooling captures the tail; the Henssge model captures the whole shape.
Worked example: a core temperature of 32°C
Using the default figures — 32°C measured, 18°C ambient, 37°C assumed starting temperature, 0.83°C per hour:
- Temperature drop: 37 − 32 = 5°C
- Estimated interval: 5 ÷ 0.83 = 6.0 hours
Now change one assumption. If the starting temperature was 36.2°C rather than 37 — well within normal variation — the estimate becomes 5.1 hours. If the true cooling rate was 0.7°C per hour rather than 0.83, it becomes 7.1 hours. Two entirely reasonable adjustments move the answer by two hours, and that is before considering clothing, body mass or whether the plateau had ended. This is the point of the exercise.
Common mistakes
- Treating a point estimate as a time of death. Real estimates are ranges with stated uncertainty.
- Assuming a normal starting temperature. Fever, exertion and hypothermia all move it substantially.
- Ignoring the plateau. The model is least reliable in the first few hours.
- Applying it in water or unusual conditions. Immersion cools a body several times faster.
- Using a surface temperature. The model assumes a deep core measurement.
Frequently asked questions
Why is this rule of thumb still taught?
Because it is memorable, it requires no equipment beyond a thermometer, and it gives a rough orientation at a scene. It is taught alongside its limitations rather than as a method to rely on. In practice, temperature-based estimation in casework uses the Henssge nomogram or equivalent, with a corrective factor and a stated confidence interval.
How accurate is temperature-based estimation generally?
Considerably less accurate than crime fiction suggests. Even the Henssge method, applied carefully under favourable conditions, carries a published confidence interval of around plus or minus 2.8 hours, and that widens substantially when conditions are unfavourable — unusual clothing, air movement, immersion, or a body found long after death. Beyond about 24 to 36 hours the method loses most of its value as the body approaches ambient temperature.
What else is used to estimate time of death?
Rigor mortis and livor mortis for the early period, gastric contents where a last meal is known, vitreous humour potassium, and forensic entomology for longer intervals. No single indicator is relied on alone. The usual approach is to establish a range from several independent methods and see where they overlap, which is what the Post-Mortem Interval Comparison tool illustrates.
Does body size matter?
Substantially. A larger body has a lower surface-area-to-mass ratio and cools more slowly; a child's body cools much faster than an adult's. Body mass is one of the primary inputs to the Henssge corrective factor for exactly this reason, and its absence is one of the largest weaknesses of a simple linear model.
Is what I enter stored?
No. Measurements are processed entirely in your browser and never transmitted or retained. Do not enter identifiable case information into any public website.
Related tools
References
- National Institute for Health and Care Excellence — clinical guidance on body temperature measurement and normal ranges
- GOV.UK — the role of the coroner and forensic pathology in death investigation
Sources are checked at publication and can change — how I choose and check references.